About

AI in Computational Arts, Music, and Games is a research group at Data Science and AI division at the Computer Science and Engineering department at Chalmers University of Technology. Our research explores the novel aesthetics and affordances of Machine Learning (ML) and Artificial Intelligence (AI) within artistic applications.

Our research encompasses AI and ML approaches such as deep generative models, supervised, unsupervised, and reinforcement learning, evolutionary approaches, multi-agent systems, and natural language processing within creative domains. These domains include multimedia applications such as music and sound art, video games, and interactive arts. In our research, the specifics of domain, problem, task, or real-world context guide our technological and algorithmic choices, rather than adhering to independent disciplinary silos. We position this interdisciplinary approach under the encompassing term, computational arts.

AI in Computational Arts, Music, and Games (AI-CompArts) contributes to the CSE department in a broader context with a unique interdisciplinary research and education at the intersection of creative domains and Artificial Intelligence. Computational Arts is an interdisciplinary field that merges principles and techniques from computer science, artificial intelligence, and digital media to create, analyze, and interact with multimedia works. The discipline involves the use of algorithms, software, and hardware to generate, manipulate, and present multimedia content in various forms, including visual art, music, sound, and multimedia installations. This field explores the novel affordances of computational methods, such as deep generative models (also referred to as generative AI), to push the boundaries of conventional art forms and create innovative interactive technology.

Conventional machine learning and artificial intelligence emphasize the end-product, or gestalt, and optimization, addressing questions such as: which architecture performs better or faster? How do we generate high-quality content such as images, videos, audio, or music? While these questions and their outcomes are relevant and significant contributions, investi- gations into AI technology processes, actors, stakeholders, diversity, inclusion, representation, and sustainability can further benefit societal discourses. To achieve this, we employ a mix of methods that combine quantitative analysis of datasets, models, algorithms, and content; with qualitative perspectives from social sciences and humanities, thereby connecting scientific knowledge to its societal impact.

Acknowledgements

The work of this research group was partially supported by the Wallenberg AI, Autonomous Systems and Software Program—Humanity and Society (WASP-HS), funded by the Marianne and Marcus Wallenberg Foundation and the Marcus and Amalia Wallenberg Foundation.